Aw, i was in favor of the separate Joseki category. @Groin brings up a nice alternative with calling it “Expected” instead. I didn’t want to speak against the majority too much though in fear of getting lynched. I’d actually prefer separating “Excellent” from book moves in order to reward creativity/skill, but that’s just me apparently. And people would definitely question how the moves are determined to be “Joseki” or “Expected” every day.
Maybe in the future if the issue comes up again we could revisit it. I can definitely appreciate the elegance of simplifying the categories too.
It would be helpful if the categories were clickable, so that you can see which moves are excellent, good, inaccurate, mistake, etc. Is this perhaps the case behind the pay wall?
That’s strange. I’m a site supporter, but when I go to a game where it only has the free AI review, such as this, I can still see the moves. Even if I open it in a private window, so I’m not logged in, I see this:
What are the definitions of the categories, are they documented anywhere? A little ? Icon linking to it or a tooltip would help. I’ve only viewed this UI on mobile so far so maybe there is already a tooltip and I just can’t see it?
I think if a site supporter does a full AI review of a game, then all people can see it. I presume that’s what happened to my game vs breakfast which let me see the per category breakdown without being a supporter. I’ve seen similar “free to all” reviews added to other interesting games like pro broadcasts.
The table of categories with count of moves per category, yes, that’s available to all in all games. But clicking the chevron next to those move counts to get a dialog listing the move numbers which made up that count is only on games with a paid review.
Good to see that the Joseki category was removed. Go is not chess
Joseki might be established and balanced locally, but not globally.
The table does not properly resize for me (Firfox). If I reduce the width of a window while the table is on the right side of the board the percentage column for white gets cut of more and more.
Something I noticed, not with a huge sample size admittedly, is that strong players and even top pros of today usually get more inaccuracies than great or good, whereas this game by the great Honinbo Shuei (white) he gets a more tapered range with greats and goods then inaccruacies in decreasing counts:
@anoek what are the definitions for the categories, is it simply something like blue move (or better!) = excellent, great is points loss <0.2, good >0.2 and < 0.5? Or some more complex combination of percentage change or other factors too?
)
